Post by thread:[EE] Looking for N-ch MOSFET TO-92

Good day to all. I'm looking for a N-channel MOSFET in a leaded package similar to TO-92 but with pinout of S-D-G. Rdson in the range of 0.1R thru 5R is good (lower is better) and Vds max of 30V or better. The 2n7000 is nearly exactly what I want, except that it has the wrong pinout (S-G-D) and the TO-92 package is a little too thick. I've been using digital transistors (DTC143ES) but this time I need a MOSFET rather than a saturated bipolar transistor. Its a long story, but I'm using it to mute a low-level audio signal. MOSFETs work very well in this type of application, so long as the negative peak audio level does not approach the turn-on voltage of the intrinsic back-body diode in the MOSFET. That is the case for this project. For What Its Worth, the DTC family of digital transistors is great! They are still available in the thin (0.075" thick) TO-92 leaded variety which makes them great for bread-boarding and vero-board and such. Any suggestions gratefully accepted! dwayne -- Dwayne Reid <spam_OUTdwaynerTakeThisOuTplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax http://www.trinity-electronics.com Custom Electronics Design and Manufacturing

Dwayne Reid wrote: > Good day to all. > > I'm looking for a N-channel MOSFET in a leaded package similar to > TO-92 but with pinout of S-D-G. Rdson in the range of 0.1R thru 5R > is good (lower is better) and Vds max of 30V or better. > > The 2n7000 is nearly exactly what I want, except that it has the > wrong pinout (S-G-D) and the TO-92 package is a little too thick. > > I've used BS170 and VN0610 also. Any I've seen use DGS (or SGD depending on view). I've one radio that uses a fet in a PCB that previously had bipolar and they just changed some parts and inserted it at right angles, the silk screen isn't even updated! In days of yore I've seen plastic adaptor used to change order of pins, or indeed just preformed and informative silkscreen. Unless this is very large volume, why worry? If the TO-92 is too fat, why not use SMD part? 2N7000, BS170, VNxxxx are all available in SMD.SOT23 (MMBF170 is SMD version of TO92 BS170) 0.1 Ohm is normally larger hexfet etc type devices TO220 or SMD with tab? The T092 parts seem to be typically 1 Ohm

I previously wrote: >Good day to all. > >I'm looking for a N-channel MOSFET in a leaded package similar to >TO-92 but with pinout of S-D-G. Rdson in the range of 0.1R thru 5R >is good (lower is better) and Vds max of 30V or better. I must be getting old or something. I went down to my bench and looked at the project I just finished last week. Guess what? It *HAS* a MOSFET that pretty much matches my description of above. Yeah - its got thick leads and a small exposed tab - but so what! I guess that I had the mindset of wanting a package like that of a 2n7000 or a DTC143ES transistor with thin wire leads. But - I don't NEED thin wire leads. Anyway, the part I used last week (and now, this week) is an IR part: IRLU120N It has Rdson of 0.185R, VDss of 100V, ID of 10A Its also relatively inexpensive: CA $0.66 in hundreds from Digikey. Just thought that I'd pass on the information. Many thanks! dwayne -- Dwayne Reid <.....dwaynerKILLspam@spam@planet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax http://www.trinity-electronics.com Custom Electronics Design and Manufacturing

At 03:15 PM 6/6/2010, Michael Watterson wrote: >I've used BS170 and VN0610 also. Any I've seen use DGS (or SGD depending >on view). > >Unless this is very large volume, why worry? Nope - this is not high volume. Couple of dozen units. Its going to be used in the little bread-board area on one of my PIC-based projects. But: the area I have to work with is extremely limited. That's because the parts associated with other PIC pins is right next to these parts. Its also got to be easy to work with so that I can hand it down to the production people so that THEY can build it. I have indeed used crossed legs on transistors before but I consciously avoid that whenever possible. >If the TO-92 is too fat, why not use SMD part? 2N7000, BS170, VNxxxx are >all available in SMD.SOT23 (MMBF170 is SMD version of TO92 BS170) >0.1 Ohm is normally larger hexfet etc type devices TO220 or SMD with >tab? The T092 parts seem to be typically 1 Ohm This area of the board is primarily through-hole. That's because it was designed to allow easy and flexible i/o configuration in projects that aren't fully defined when the board layout is done. The layout is indeed conducive to using 0805 or 0603 SMT 2-leaded parts (resistors or capacitors) but not at all suited for SOT-23 packages. I really don't like going full-steam into projects where changes are probably going to occur during and after the boards are built but sometimes time-lines just don't allow me the luxury of completing the development before production starts. Yeah - means taking stuff apart to make changes. But the customer who requests those changes pays for the re-work. In this case, this box is part of a much larger project that is mostly shipped. The customer discovered some real problems that he hadn't anticipated and now I have to make it work the way he wants (as opposed to what he asked for). Its times like this when I'm REALLY glad that I include that flexible i/o area in many of my PIC projects - its what is making these changes much less labor intensive and painful than they otherwise would be. dwayne -- Dwayne Reid <dwaynerKILLspamplanet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax http://www.trinity-electronics.com Custom Electronics Design and Manufacturing

>> Since this is obviously a one-off (else thru hole would be silly), ... > Most of non one-offs on the market, as it had been discussed already > here, are assembled on single-sided PCBs, many of them (if not most) > are "thru hole". > So, the question: Are all of those developers silly idiots? :-) Hat: CASE contextual. 1. Dwayne's sig line strongly suggests that he is in Canada, and his long list history is known to the respondent who used the term 'silly' so it would be preferred (sez Occam) to consider that comment in that context. In many cases the use of the term 'silly' would be correct, silly being readily taken to have the common usage meaning of "not apparently anything like optimal". 2. "Silly idiots" carries quite a different sense, usually, to most list members than "silly" does in the above context. This is an important point. If said respondent had said " ... silly idiot ..." then his response would be understood quite differently than it otherwise are reasonably is. This is important, he repeated :-). So 3. Answering your explicit question and taking "silly idiots" to mean "behaving in an extremely illogical and sub optimal manner" then: > Most of non one-offs ... here, are assembled on single-sided PCBs, many of them (if not most) > are "thru hole". > So, the question: Are all of those developers silly idiots? :-) Important to know: - By "here" do you mean- where Dwayne is or where you are? If "where you are" then, where is that ? Dwayne had explained his criteria and in that context the answer would be "no", and nor was that implied by the respondent. If you are referring to where you are, then they may or may not be BUT if they are involved in long term commercial development then we can assume that they are driven by market forces so again we expect the answer to be "no". It would, presumably, be some halcyon locale with low labor rates relative to material costs, untroubled by the need for compactness, speed of manufacture, ease of achieving high quality, high availability of parts or competitiveness with people who have access to such. Such places do still exist. This could be eg cottage industry level manufacture in eg China or other Asian areas, probably also India and also some parts of Africa. I'd be genuinely interested in knowing if that was what you meant and, if so, where 'here" is. . R

At 06:18 AM 6/7/2010, Olin Lathrop wrote: >Since this is obviously a one-off (else thru hole would be silly), Actually, most of the stuff we build is through-hole. Our customers don't consider it silly at all. Our highest-volume products use 1 or 3 AZ2150 relays and anywhere from 1 to 8 JS1 relays on each board. The AZ2150 relays are similar to the Potter & Brumfield T90 relays - good for up to 277Vac and up to 35A. The terminal blocks are all rated for 20A or better. NONE of that stuff is suited for SMT - not at the current levels they are expected to work at. Although the remainder of the board could be SMD, we choose to build the entire board as through-hole. There just isn't that much other stuff on the board. That may change in the future, though. What won't change is that all of those through-hole components will still be wave-soldered when the boards are built. Something to keep in mind, Olin. I think that most of your design work is for very high-volume customers. SMT is probably the only way you can go for most or all of that stuff. Our market niche caters to much smaller volumes. Yeah - we build about 10,000 boards per year for one customer (3 different designs). But we also make things one at a time for those customers who need them. Through-hole is often the easiest / fastest way to build those very low volume projects. One thing in common with most of our customers - they expect a very long lifetime for their products. The HVAC stuff needs to run trouble-free for at least 10 years. Some of the stuff we build for Dow Chemical is expected to work for 20 years. dwayne -- Dwayne Reid <.....dwaynerKILLspam.....planet.eon.net> Trinity Electronics Systems Ltd Edmonton, AB, CANADA (780) 489-3199 voice (780) 487-6397 fax http://www.trinity-electronics.com Custom Electronics Design and Manufacturing

>>> Since this is obviously a one-off (else thru hole would be silly), ... > >> Most of non one-offs on the market, as it had been discussed >> already here, are assembled on single-sided PCBs, many of >> them (if not most) are "thru hole". > >> So, the question: Are all of those developers silly idiots? :-) > > Hat: CASE contextual. > > 1. Dwayne's sig line strongly suggests that he is in Canada, > and his long list history is known to the respondent who used > the term 'silly' so it would be preferred (sez Occam) to consider > that comment in that context. Your manipulation with "context" word is hard to understand. Yes, I know he is in Canada. Just recall his post 3 months ago: > Most of our high-volume stuff (4K to 10K pieces per > year) is single-sided just because of that cost savings. "Single-sided" means that most of the stuff on the PCB will be "thru hole". What's wrong with that "context"? Even Olin agreed that "single sided PCBs are majority" 3 months ago: >> How am I expected to come to the conclusion that >> single sided PCBs are majority? > > Take a look under the dashboard of your car, for example. A lot of people here worked real hard to convince me that "single sided PCBs are majority" 3 months ago. (Still "single-sided" means that most of the stuff on the PCB will be "thru hole"). Remember M. Adam Davis's post: > All of the items that are created millions in a > manufacturing run are, if at all possible, done on a single > sided PCB because it saves significant money over the > entire batch. Yes, it's only pennies extra per board, but > that adds up when you manufacture 10 million of them. Now the same folks are trying to convince me back, that the concept is silly: >> Since this is obviously a one-off (else thru hole would be silly), ... You guys need to stick to one side, no need to bounce back and forth. {Quote hidden}> Important to know: > > - By "here" do you mean- where Dwayne is or where > you are? If "where you are" then, where is that ? > > Dwayne had explained his criteria and in that context > the answer would be "no", and nor was that implied by > the respondent. > > If you are referring to where you are, then they may or > may not be BUT if they are involved in long term > commercial development then we can assume that > they are driven by market forces so again we expect the > answer to be "no".It would, presumably, be some > halcyon locale with low labor rates relative to material > costs, untroubled by the need for compactness, > speed of manufacture, ease of achieving high quality, > high availability of parts or competitiveness with > people who have access to such. Such places do still > exist. This could be eg cottage industry level manufacture > in eg China or other Asian areas, probably also > India and also some parts of Africa. > > I'd be genuinely interested in knowing if that was what > you meant and, if so, where 'here" is. I'm afraid, your over verbose response (did you over verbose it on purpose?) is impossible to understand. Dwayne clearly stated 3 months ago "Most of our high-volume stuff (4K to 10K pieces per year) is single-sided just because of that cost savings." That's it, he calculated and found "because of cost savings." No need to call "China or other Asian areas, probably also India" "some halcyon locale ". They contribute almost all PCB world assembly from which, as we agreed 3 months ago, "single sided PCBs are majority".

Marechiare wrote: > Most of non one-offs on the market, as it had been discussed already > here, are assembled on single-sided PCBs, many of them (if not most) > are "thru hole". Single sided does not necessarily imply thru hole. The two are independent. > So, the question: Are all of those developers silly idiots? :-) There are still some applications where thru hole is required. For example, thru hole connectors can withstand more force and often have a smaller footprint because the leads and the connector body can be overlapped in terms of board area. Another example is extreme temperature cycling together with the cheapest possible boards. The difference in thermal expansion coefficients will stress the solder joint, which can eventually fail from the repeated stress cycles. For thru hole, the leads just flex a little bit, which results in less stress. ******************************************************************** Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products (978) 742-9014. Gold level PIC consultants since 2000.

Dwayne Reid wrote: > Actually, most of the stuff we build is through-hole. Our customers > don't consider it silly at all. Would they know one way or the other? > Our highest-volume products use 1 or 3 AZ2150 relays and anywhere > from 1 to 8 JS1 relays on each board. The AZ2150 relays are similar > to the Potter & Brumfield T90 relays - good for up to 277Vac and up > to 35A. The terminal blocks are all rated for 20A or better. > > NONE of that stuff is suited for SMT - not at the current levels they > are expected to work at. Sure, big heavy and high power stuff needs to be mounted differently. However, the discussion was about a TO-92 transistor. That is neither heavy nor has high current requirements. > Although the remainder of the board could be SMD, we choose to build > the entire board as through-hole. There just isn't that much other > stuff on the board. Have you actually priced it out to see whether it makes sense for the small electronic stuff to be thru hole or SMD? If you build enough of them with a competent manufacturer, it's probably cheaper to run a SMD pass, then do the manual stuffing and wave soldering of the thru hole parts. > Our market niche caters to much smaller volumes. Yeah - we build > about 10,000 boards per year for one customer (3 different > designs). That's plenty high enough volume for automated pick and place to be cost effective. > But we also make things one at a time for those customers > who need them. Through-hole is often the easiest / fastest way to > build those very low volume projects. If you're doing everything manually, the labor cost is going to dominate. Don't assume that even with full manual that thru hole is quicker than SMD. It may or may not be the case, depending on board particulars. I know that it takes me longer to build a thru hole board than a equivalent SMD board with ordinary electronic jelly bean components. > One thing in common with most of our customers - they expect a very > long lifetime for their products. The HVAC stuff needs to run > trouble-free for at least 10 years. Some of the stuff we build for > Dow Chemical is expected to work for 20 years. Sure, but that has nothing to do with thru hole versus SMD. Certain types of failures, like cold solder joints, are pretty much a thru hole only problem. I've heard one source claim that SMD has better long term reliability because the types of failures that occur with SMD are more likely caught during production. I'm not sure how true that is, but I certainly wouldn't just blindly assume thru hole is better for quality than SMD. ******************************************************************** Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products (978) 742-9014. Gold level PIC consultants since 2000.

> Single-sided PCBs don't make sense if are a little cheaper than double > sided because double sided offer considerable PCB space saving. To be > seriously cheaper single-sided PCBs have to be manufactured from very > cheap materials. Such stuff can't hold copper stripes with SMT > components reliably after heated. That's my version why I don't see > many full SMT single sided PCB. Perhaps there are other reasons as > well. That is not what my video recorder tells me. It has single sided phenol base PCBs, with smd components for most things, except for electrolytic capacitors, which apparently are cheaper when PCB mount leaded versions are used. There are also many jumper wires, which give the impression of being machine cut and bent as they are fitted. -- Scanned by iCritical.

>>> (Still "single-sided" means >>> that most of the stuff on the PCB will be "thru hole"). >> I don't know where you got that from. > Single-sided PCBs don't make sense if are a little cheaper than double > sided because double sided offer considerable PCB space saving. To be > seriously cheaper single-sided PCBs have to be manufactured from very > cheap materials. Such stuff can't hold copper stripes with SMT > components reliably after heated. That's my version why I don't see > many full SMT single sided PCB. Perhaps there are other reasons as > well. My solar lights are SMD DS. Somebody copied two models of our lights, rolled the mechanicals and body moldings together into a hybrid, reduced the cost of some parts, did some mechanical 'improvement' (while missing some crucial points), copied part of the electronics and produced a simplified subset version. SMD board except for a 1 Watt through hole resistor mounted on copper side. All the rest SMD but wholly single sided. Photos of their PCB sometime maybe. SMD + SS. The real thing - one version 1st version www.greenfeet.com/mmGF/Images/bogo-orange-actual.jpg 2nd version Accessory not included - http://public.fotki.com/RussellMc/atw/bogo/sl2africa01/solarlight.html http://public.fotki.com/RussellMc/atw/bogo/sl2africa01/solarlightcharge.html Clone hybrid (unauthorised) http://hitidesolar.en.ecplaza.net/20.asp PCB. One of ours DS - mainly SMD - cost aspects result in some TH. Mainly Al ecaps and inductor here. http://public.fotki.com/RussellMc/atw/bogo/sn2pcb/sn2pcbdsc00714r.html#media Russell

>> Single-sided PCBs don't make sense if are a little cheaper than double >> sided because double sided offer considerable PCB space saving. To be >> seriously cheaper single-sided PCBs have to be manufactured from very >> cheap materials. Such stuff can't hold copper stripes with SMT >> components reliably after heated. That's my version why I don't see >> many full SMT single sided PCB. Perhaps there are other reasons as >> well. > That is not what my video recorder tells me. It has single sided phenol > base PCBs, with smd components for most things, except for electrolytic > capacitors, which apparently are cheaper when PCB mount leaded versions > are used. Yes. Substantially so when you get above a few uF at low voltage. Based on all up cost caps + inductor on our large lights is TH. Small "Mini" has TH ecaps and inductor is SMD. TH inductor may have been cheaper !. Wiring connectors are usually TH. You can get small SMD connectors but copper peel risk is greater. Connectors may have fingers around board edge to (try to) stop this. Data point: Find old SIM RAM. They have a nice hole on one end that takes a keyring ring. Add ring and keys. Carry in pocket. Inspect visually every now and then and see how much RAM capacity remains. ICs will fall off surprisingly quickly under migration pressure of being in pocket until you have a bare PCB. Shame - quite a fun keyring. ICs probably fall off more rapidly in my pocket than in most people's :-). Clanking noises have been heard when walking. Holes used to happen. Better choice of pocket linings helps. R . There are also many jumper wires, which give the impression of > being machine cut and bent as they are fitted.

> Clone hybrid (unauthorised) > > http://hitidesolar.en.ecplaza.net/20.asp > > PCB. One of ours DS - mainly SMD - cost aspects result in some TH. Yes, one could expect anything from pirates. Perhaps I should have said "Still "single-sided" means that for most PCBs most of the stuff on the PCB will be "thru hole" instead of: >> (Still "single-sided" means >> that most of the stuff on the PCB will be "thru hole"). Of course, there are classes of applications that do make sense using SMT on single sided PCBs, for example cheap SMT LED panels. But for majority apps, I believe, PCB space saving and increased reliability would bring second copper layer and through hole plating cost effective. If I were to buy the gadget I would definitely prefer your original "SMD DS" to the pirated one. (Haven't I just earned an indulgence in advance from the admin for my next posts? :-)